Coursera Spring Sale
40% Off Coursera Plus Annual!
Grab it
Explore the nuanced realities of artificial intelligence beyond its promise to understand its inherent limitations and discover how these constraints can become competitive advantages in this conference presentation by Prof. Mario Trapp from Fraunhofer Institute for Cognitive Systems IKS. Learn why safe AI remains elusive in physical and cyber-physical systems, and understand how AI quality limitations present strategic opportunities, particularly for German industry. Examine the triarchic intelligence framework that combines functional, adaptive, and cooperative intelligence to build resilient systems, and gain real-world insights into embedding AI within traditional safety engineering systems. Discover the critical importance of human-in-the-loop approaches and how humans remain essential components of safe, intelligent systems through practical examples ranging from detecting people in industrial settings to adaptive driving systems and human-AI collaboration. Understand safety as a system property where AI safety must be engineered at the system level, explore adaptive supervisors that balance risk and utility instead of overly conservative safety systems, and learn why human-AI collaboration is essential rather than burdensome. Gain insights into education and accountability for training professionals to understand AI decisions and enable responsible oversight, while discovering a realistic perspective on AI strategy grounded in practical implementation rather than hype, along with a roadmap for building AI in safety-critical systems without compromising risk management.
